KMS XINJIANG INSTITUTE OF ECOLOGY AND GEOGRAPHY，CAS
| 基于遥感-生态模型的中亚草地生态系统净初级生产力对气候变化和放牧的响应研究 | |
| 叶辉 | |
| Subtype | 博士 |
| Thesis Advisor | 罗格平 |
| 2018-06-05 | |
| Degree Grantor | 中国科学院大学 |
| Place of Conferral | 新疆乌鲁木齐 |
| Degree Discipline | 理学博士 |
| Keyword | 中亚干旱区 遥感 植被净初级生产力 放牧 气候变化 remote sensing Defoliation Formulation Model net primary production grazed land spatial-temporal patterns Central Asia |
| Abstract | 亚洲中部干旱区（包括中亚五国地区（以哈萨克斯坦、乌兹别克斯坦、吉尔吉斯斯坦、土库曼斯坦以及塔吉克斯坦）以及我国新疆等地区， 本研究中简称“中亚干旱区”）是中纬度最具代表性的干旱区之一，作为新丝绸之路核心区，对全球气候暖化、降水格局（强度和周期）变化、 以及大气成份变化极其敏感。过去几千年， 放牧一直是干旱半干旱区主要的人类活动。近 50 年来，中亚干旱区放牧强度持续增加，过度放牧已成为草地生态系统普遍存在的现象， 这使得原本脆弱的草地生态系统退化日益加剧，在气候变化背景下尤为脆弱。草地净初级生产力是食草动物的主要消费来源， 对干旱区草地净初级生产力（NetPrimary Production， NPP） 进行量化， 揭示放牧过程对干旱草地生态系统 NPP的影响程度，提出合理的干旱区草地放牧制度，促进草地生态系统的可持续发展具有重要的理论与现实意义。基于植被指数的遥感模型已广泛运用于区域尺度模拟草地净初级生产力的模拟，在干旱区的运用充分发挥了其高时空分辨率的优势。 然而， 遥感模型是基于动物啃食后的草地残余植被信息模拟的草地净初级生产力，忽略了人类干扰诸如放牧活动消耗的部分， 并非真实状态下的草地净初级生产力。鉴于此，针对其独特的生理生态机制和以放牧为主的人类活动干扰，本研究将GLOPEM-CEVAS 遥感模型应用于中亚干旱区，在模型本地参数化的基础上，通过多种模型对比和实测数据来评估其适用性和可靠性， 同时通过加入落叶方程模型， 考虑放牧过程消耗的 NPP 后，评价放牧对于亚洲中部干旱区草地生态系统 NPP 的影响。 获得的主要结论如下：（1） 对中亚干旱区草地生态系统 NPP 模拟结果表明， 与生态模型结果的对比，GLOPEM-CEVAS 模型由于忽略了放牧过程，其结果相对生态模型较低。。（2） 在放牧条件下 GLOPEM-CEVAS 模型模拟的 36 个放牧样点草地 NPP 与实测数据显著相关（AdjR2 = 0.80, p < 0.01)，模型能够解释草地实测值的 80%。在无放牧条件下， GLOPEM-CEVAS 模型模拟了 4 个无放牧活动的采样点，草地平均 NPP 为 356.44 ± 86.78 gC m-2，与实测值 370.23 ± 41.88 gC m-2 接近，GLOPEM-CEVAS 模型能较好的模拟新疆草地 NPP。（3） 未考虑放牧活动时，基于 GLOPEM-CEVAS 模型模拟的 1982-2011 年新疆草地年均 NPP 为 121.05 gC m-2 yr-1，草地 NPP 波动增加但不显著(AdjR2 = 0.06,p > 0.05)，其年际波动方式和年均降水较为相似（AdjR2 = 0.24, p < 0.01），与气温相关性不明显； 考虑放牧过程加入落叶方程模型后，模型模拟的新疆草地年均真实 NPP 为 179.41 gC m-2 yr-1，草地 NPP 持续增加显著（slope = 1.18 gC m-2yr-1,AdjR2 = 0.63, p < 0.01)，其年际 NPP 和年均气温显著相关（AdjR2 = 0.42, p < 0.01），与降水相关性不明显， 放牧消耗的平均 NPP 占草地 NPP 的 29.06%；新疆草地实际 NPP 逐年增加且由温度主导，其增加部分被放牧活动所消耗，忽略放牧活动会掩盖这一事实，这会对气候变化对新疆草地的影响带来错误的理解。（4） 忽略放牧过程的影响下， GLOPEM-CEVAS 模型模拟的中亚干旱区草地年均 NPP 为121.05 gC m-2 yr-1，草地 NPP 波动增加显著(slope = 1.02 gC m-2 yr-1,AdjR2 = 0.43, p < 0.01)，最大值和最小值分布是 1996 年的 167 gC m-2 和 1984 的121 gC m-2，与年际降水的相关性不显著 (AdjR2 = 0.12, p = 0.03)， 与年际气温相关性显著(AdjR2 = 0.44, p < 0.01)；考虑放牧活动后，中亚草地 NPP 持续增加(slope= 2.7 gC m-2 yr-1, AdjR2 = 0.86, p < 0.01)， 2000 以后的增加趋势尤为明显，放牧活动对中亚干旱区草地 NPP 的消耗极为显著，与年际气温同样相关(AdjR2 = 0.49, p< 0.01)，与年际降水不相关。（5） 在空间上，考虑放牧活动后中亚草地 NPP 空间分布从大到小依次为：哈萨克斯坦北部草原（257 gC m-2yr-1）、天山地区（208 gC m-2yr-1） 、中部荒漠草原带（105 gC m-2yr-1） 。从区域上比较，哈萨克斯坦北部草原 NPP 最大，中部荒漠草原 NPP 最小，哈萨克斯坦北部草原是放牧消耗 NPP 的主要区域，放牧使得草地 NPP 平均减少了 99 gC m-2yr-1， 占 NPP 的 20%-25%。 |
| Other Abstract | As the core area of the New Silk Road, Central Asia (including the five Central Asiancountries (Kazakhstan, Uzbekistan, Kyrgyzstan, Turkmenistan, and Tajikistan) andChina's Xinjiang and other regions, which are sensitive to global climate change (etc.changes in precipitation patterns (intensity and cycle), and changes in atmosphericcomposition). For the past few millennia, grazing has been the main human activityin arid and semi-arid regions and the grazing intensity in the arid regions of CentralAsia has continued to increase in the past 50 years. Overgrazing has become awidespread phenomenon in grassland ecosystems that is already vulnerable, and isparticularly vulnerable to climate change and it has increasingly become more andmore deteriorating. Net primary productivity (NPP) of grassland is the main source ofconsumption of herbivores. Quantifying net primary production of grassland is a keycomponent of production management in arid regions which manifests the degree ofimpact of grazing on NPP in arid grassland ecosystem. It has a greatly importanttheoretical and practical significance to promote the sustainable development ofgrassland ecosystem.The remote sensing model based on vegetation index has been widely used tosimulate the grassland's net primary productivity at the regional scale. The applicationof the remote sensing model in the arid region fully exploited its advantages of highspatial and temporal resolution. However, the remote sensing model is based on thenet primary productivity of grassland modeled after the animal's foraging grasslandresidual vegetation information which neglected human interference such as theconsumption of herbivores, not the real grassland primary productivity. Therefore, forthe unique physiological and ecological mechanisms and disturbances ofgrazing-based human activities, GLOPEM-CEVAS, a kind of remote sensing modelis applied to the arid region of Central Asia. Based on the local parameterization ofthe model and model comparision, the measured data were used to evaluate its applicability and reliability. At the same time, a defoliation formulation model (DFM)based on RS is developed to evaluate the extent of underestimated NPP to consideratethe NPP consumed by the grazing process. After that, we analyzed response of netprimary productivity to climate change and human activities on grassland based onremote sensing model in Central Asia. The main findings are as follows:(1) The results of NPP simulation of the grassland ecosystem in the arid region ofCentral Asia show that, compared with the results of the ecological model, theGLOPEM-CEVAS simulation results are relatively low due to the grazing process areneglected.(2) The grassland NPP was significantly correlated with the observed data (AdjR2 =0.80, p < 0.01) in the 36 grazing plots simulated by the GLOPEM-CEVAS modelunder grazing conditions, and the model can account for 80% of the actual grasslandmeasurements. Under the no-grazing conditions, the GLOPEM-CEVAS modelsimulated four sampling sites without grazing activity. The average NPP of thegrassland was 356.44 ± 86.78 gC m-2, which was close to the measured value of370.23 ± 41.88 gC m-2. The GLOPEM-CEVAS model was able to simulate Xinjianggrassland NPP.(3) When grazing is not considered, the annual NPP of grassland in Xinjiang duringthe period from 1982 to 2011 based on the GLOPEM-CEVAS model is 121.05 gCm-2 yr-1. The grassland NPP fluctuates but does not increase significantly (AdjR2 =0.06, p> 0.05). The interannual fluctuation pattern and average annual precipitationare similar (AdjR2 = 0.24, p < 0.01), and the correlation with temperature is notobvious. After considering the grazing process to join the defoliation equation model,the simulated average annual NPP of Xinjiang grassland is 179.41. gC m-2 yr-1, thegrassland NPP continued to increase significantly (slope = 1.18 gC m-2yr-1, AdjR2 =0.63, p <0.01), and its annual NPP was significantly correlated with the annual meantemperature (AdjR2 = 0.42, p < 0.01), the correlation with precipitation is not obvious,and the average NPP consumed by grazing accounts for 29.06% of grassland NPP;actual grassland NPP in Xinjiang increases year by year and is dominated bytemperature, and its increase is consumed by grazing activities. Ignoring grazing willobscure this fact. This will lead to a misunderstanding of the effects of climate changeon grassland in Xinjiang.(4) Neglecting the influence of grazing process, the annual NPP of grassland in thearid region of Central Asia simulated by GLOPEM-CEVAS model is 121.05 gC m-2yr-1, and the NPP fluctuation in grassland is significantly increased (slope = 1.02 gCm-2 yr-1 (AdjR2 = 0.43, p < 0.01), the maximum and minimum distributions were 167gC m-2 in 1996 and 121 gC m-2 in 1984, and the correlation with interannualprecipitation was not significant (AdjR2 = 0.12, p = 0.03), correlated significantlywith interannual air temperature (AdjR2 = 0.44, p < 0.01); after considering grazing,NPP in Central Asian grassland continued to increase (slope = 2.7 gC m-2 yr-1, AdjR2= 0.86, p < 0.01 The trend of increase after 2000 is particularly obvious. Grazingactivity has a very significant depletion of NPP in the arid regions of Central Asia,and is also related to interannual temperatures (AdjR2 = 0.49, p < 0.01), and is notrelated to interannual precipitation.(5) At the spatial scale, considering the grazing activities, the spatial distribution ofNPP in Central Asia grasslands was ranked as follows: Northern Kazakhstangrassland (257 gC m-2yr-1), Tianshan area (208 gC m-2yr-1), The middle desert steppezone (105 gC m-2yr-1). In terms of regional comparison, the northern grassland NPPin Kazakhstan has the largest NPP and the central desert grassland has the smallestNPP. The northern Kazakhstan grassland is the main area for grazing consumption ofNPP. Grazing results in an average NPP reduction of 99 gC m-2yr-1, accounting for 20%-25%of NPP. |
| Subject Area | 地图学与地理信息系统 |
| Language | 中文 |
| Document Type | 学位论文 |
| Identifier | http://ir.xjlas.org/handle/365004/14941 |
| Collection | 研究系统_荒漠环境研究室 |
| Affiliation | 中国科学院新疆生态与地理研究所 |
| First Author Affilication | 中国科学院新疆生态与地理研究所 |
| Recommended Citation GB/T 7714 | 叶辉. 基于遥感-生态模型的中亚草地生态系统净初级生产力对气候变化和放牧的响应研究[D]. 新疆乌鲁木齐. 中国科学院大学,2018. |
| Files in This Item: | There are no files associated with this item. | |||||
Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Edit Comment